INDEX

Explanations

phrases related to technology or specific product names

oai_token-act-pair · gpt-3.5-turbo

titles or names of specific entities related to various subjects, such as films, organizations, and characters

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etheless

-0.78

CONCLUS

-0.69

BUT

-0.63

Done

-0.62

 anyway

-0.61

Nonetheless

-0.61

Likewise

-0.60

dding

-0.59

cluding

-0.59

Tes

-0.59

POSITIVE LOGITS

 consists

1.42

 refers

1.33

 comprises

1.28

 represents

1.21

 originated

1.12

 combines

1.12

 specializes

1.10

is

1.08

 involves

1.07

 revolves

1.05

Activations Density 0.518%

phrases related to technology or specific product names

titles or names of specific entities related to various subjects, such as films, organizations, and characters

No Comments

No Known Activations